Please turn on javascript in your browser to play chess.
Only Chess Forum

Only Chess Forum

  1. Subscriber Mutt n Stu
    K.I.N.G.
    02 Apr '08 13:17
    I was analysing a recent (completed) game of mine to see where I had gone wrong. Given that I was mated in 12 moves you could reword that to say "horribly wrong". Anyhow, the Sicilian Alapin variation was played for the first couple of moves and I figured out pretty easily where I went wrong. Pretty obvious really when I look at it afterwards.

    The interesting part was when I ran this game through ChessMaster and ChessBase Light. The game had got to Black's second move (1. e4 c5 2. c3) and I played Nf6 - as per the book line and (therefore not surprisingly) one of the 2 most popular responses (the other being d5). However, CM and Fritz suggested either Nc6 or d5 which surprised me.

    Now, is this just a case of an engine thinking like an engine? I would have thought that a program like Chessmaster which has a database of openings would take well researched and established theoretical lines into account in its analysis?
  2. Standard member wittywonkaonline
    Chocolate Expert
    02 Apr '08 13:38
    Originally posted by Mutt n Stu
    I was analysing a recent (completed) game of mine to see where I had gone wrong. Given that I was mated in 12 moves you could reword that to say "horribly wrong". Anyhow, the Sicilian Alapin variation was played for the first couple of moves and I figured out pretty easily where I went wrong. Pretty obvious really when I look at it afterwards.

    The inte ...[text shortened]... gs would take well researched and established theoretical lines into account in its analysis?
    Well, 2. ... d5 is certainly a reasonable choice against the 2. c3 Variation, but 2. ... Nc6 is indeed strange. Of course, it is playable, but it certainly does concede the center rather quickly.
  3. Standard member Korch
    Chess Warrior
    02 Apr '08 13:54
    Originally posted by Mutt n Stu
    I was analysing a recent (completed) game of mine to see where I had gone wrong. Given that I was mated in 12 moves you could reword that to say "horribly wrong". Anyhow, the Sicilian Alapin variation was played for the first couple of moves and I figured out pretty easily where I went wrong. Pretty obvious really when I look at it afterwards.

    The inte ...[text shortened]... gs would take well researched and established theoretical lines into account in its analysis?
    Engines without databases are bad advisors in opening. Books are much better.
  4. 02 Apr '08 13:58
    Was the program's Opening Book turned off?
  5. Subscriber Mutt n Stu
    K.I.N.G.
    02 Apr '08 14:22
    Originally posted by Maxwell Smart
    Was the program's Opening Book turned off?
    Don't know is there is a setting for that or not. Chessmaster has an awful manual and my laptop's display/resolution has some problem with Chessmaster's menu (I can't see the menu title unless connected to an external monitor but I can see the menu drop downs if I hover / click over the place where I know they are). Anyhow, I haven't seen any Opening Book setting but then again I haven't really been on the lookout for it. I'll take a look and see if I can find something.
  6. 02 Apr '08 14:49 / 1 edit
    Originally posted by Mutt n Stu
    Don't know is there is a setting for that or not. Chessmaster has an awful manual and my laptop's display/resolution has some problem with Chessmaster's menu (I can't see the menu title unless connected to an external monitor but I can see the menu drop downs if I hover / click over the place where I know they are). Anyhow, I haven't seen any Opening Book ...[text shortened]... en't really been on the lookout for it. I'll take a look and see if I can find something.
    first tip: do not chess engines for analysis in the first moves of an opening. unnecessarily favor pawns at e4 and d4, and knights in c3 and f3 in the opening.

    second tip: do not use chessmaster for any kind of analysis. never. ever.
  7. Standard member wormwood
    If Theres Hell Below
    02 Apr '08 15:18
    Originally posted by Korch
    Engines without databases are bad advisors in opening. Books are much better.
    I think the same applies to any position. the opening just is the only one which is so thoroughly analysed that we know when the engine is claiming nonsense.
  8. 02 Apr '08 15:20 / 1 edit
    Originally posted by Korch
    Engines without databases are bad advisors in opening. Books are much better.
    Doesn't an engine advise 5.Nxf7?! in the Traxler.
    Any specialist book (or good database) will tell you that 5.Bxf7+!? is better for White.
  9. Standard member Korch
    Chess Warrior
    02 Apr '08 15:26
    Originally posted by wormwood
    I think the same applies to any position. the opening just is the only one which is so thoroughly analysed that we know when the engine is claiming nonsense.
    Well, in many books, written before engine era, analysis given by humans are refuted by engine.
  10. Standard member Korch
    Chess Warrior
    02 Apr '08 15:27 / 1 edit
    Originally posted by Squelchbelch
    Doesn't an engine advise 5.Nxf7?! in the Traxler.
    Any specialist book (or good database) will tell you that 5.Bxf7+!? is better for White.
    Yes - even the best engines may not understand positions in gambits like Traxler (and also Latvian gambit).
  11. Subscriber Mutt n Stu
    K.I.N.G.
    02 Apr '08 15:40
    Originally posted by diskamyl
    second tip: do not use chessmaster for any kind of analysis. never. ever.
    What do you recommend? I like chessmaster for the tutorials but find its analysis quite cumbersome so looking to get some new software (preferably something that has a Mac version).

    I see is some software you can load other engines? How does this work? E.g. I assume something like ChessBase can use a number of different engines and ChessBase is just the shell / skin / whatever you want to call it and then you can load whatever engine you want to run the actual analysis, say Fritz. (assuming whatever you choose is compatible).

    If the above is correct then I guess there are 2 questions: what program should i use and what engine?
  12. Standard member wormwood
    If Theres Hell Below
    02 Apr '08 15:49
    Originally posted by Korch
    Well, in many books, written before engine era, analysis given by humans are refuted by engine.
    that's also true of course. but aren't those errors mostly tactical oversights? - fortunately we have both approaches to combine and complement each other's weaknesses...
  13. Standard member Korch
    Chess Warrior
    02 Apr '08 16:20
    Originally posted by wormwood
    that's also true of course. but aren't those errors mostly tactical oversights? - fortunately we have both approaches to combine and complement each other's weaknesses...
    You should not underrate importance of tactics. As Tigran Petrosian once have said - tactic is the most important thing.
  14. 02 Apr '08 16:25
    This may be a little off topic, but it's a funny story. thgibbs and I were working on a new opening. We agreed that we would use books and/or databases to learn this opening. Really, our main purpose was to get through the opening to concentrate our efforts on the middlegame. I was not paying attention and just played what was reported as a main line and suddenly discovered that I was a piece down! What we call "book" has been laid down by grandmaster research; some popular line in a database may be nothing more than trash.

    I don't know anything about the discussion on comparing engines with book, but I would assume from my elementary knowledge of Crafy that if the opening book was filled with the right stuff the engine would follow book from a table search.

    Years ago I built an opening book of my own games thinking I would learn to find my own weaknesses. I did. I discovered my weakness was everything. That experiment was a total flop and not worth the time it took to construct. The experience was similar to the example in the first paragraph.
  15. 02 Apr '08 17:02 / 4 edits
    Originally posted by Mutt n Stu
    What do you recommend? I like chessmaster for the tutorials but find its analysis quite cumbersome so looking to get some new software (preferably something that has a Mac version).

    I see is some software you can load other engines? How does this work? E.g. I assume something like ChessBase can use a number of different engines and ChessBase is just t ...[text shortened]... above is correct then I guess there are 2 questions: what program should i use and what engine?
    if you have any intention of buying software, wait for Rybka 3.0 to be released. it's undisputedly the best chess engine available, and 3.0 will also be in it's own GUI. (http://www.rybkachess.com). It will probably go out in a few months.

    until then / or if you have no intention of buying any software, there are many free engines a lot better Chessmaster's engine (theking), some are fruit and toga. just google these and/or visit www.superchessengine.com.

    you can use the engines mentioned above and most of the ones in superchessengine.com in chessbase. just click Engine->Create UCI Engine... and you'll figure out the rest.

    you can analyse with the engine using "add kibitzer" from the same engine menu. you will not be able to believe yourself how you managed to analyse games with that CM interface.

    (I love CM by the way, but only for tutorials and playing).

    -
    I don't know about the Mac thing though. I had heard shredder had a mac version. it's always been one of the top 5 engines. (www.shredderchess.com)